MOTSynth: How Can Synthetic Data Help Pedestrian Detection and Tracking?

Fabbri, Matteo; Brasó, Guillem; Maugeri, Gianluca; Cetintas, Orcun; Gasparini, Riccardo; Ošep, Aljoša; Calderara, Simone; Leal-Taixé, Laura; Cucchiara, Rita

doi:10.1109/iccv48922.2021.01067

Cited by 87 publications

(38 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[253] introduces a large-scale synthetic data engine named MOTX to generate synthetic data for MOT training and achieve competitive results with the trackers trained using real data. In addition, [254] shows that better performance can be achieved with synthetic data on some widely known trackers such as Tracktor [15], Track R-CNN [37], Lift T [255], MPNTrack [6], and CenterTrack [12]. These works prove the efficiency of utilizing synthetic data for the MOT task.…”

Section: Training With Synthetic Datamentioning

confidence: 87%

See 1 more Smart Citation

Recent Advances in Embedding Methods for Multi-Object Tracking: A Survey

Wang¹,

Song²,

Hwang³

2022

Preprint

View full text Add to dashboard Cite

Multi-object tracking (MOT) aims to associate target objects across video frames in order to obtain entire moving trajectories. With the advancement of deep neural networks and the increasing demand for intelligent video analysis, MOT has gained significantly increased interest in the computer vision community. Embedding methods play an essential role in object location estimation and temporal identity association in MOT. Unlike other computer vision tasks, such as image classification, object detection, re-identification, and segmentation, embedding methods in MOT have large variations, and they have never been systematically analyzed and summarized. In this survey, we first conduct a comprehensive overview with in-depth analysis for embedding methods in MOT from seven different perspectives, including patch-level embedding, single-frame embedding, cross-frame joint embedding, correlation embedding, sequential embedding, tracklet embedding, and cross-track relational embedding. We further summarize the existing widely used MOT datasets and analyze the advantages of existing state-of-the-art methods according to their embedding strategies. Finally, some critical yet under-investigated areas and future research directions are discussed.

show abstract

Section: Training With Synthetic Datamentioning

confidence: 87%

“…Since fully annotated real video data is hard to collect, many works [229], [253], [254] try to utilize synthetic data for training trackers. The annotations such as box, mask and track ID, can be naturally obtained from synthetic data.…”

Section: Training With Synthetic Datamentioning

confidence: 99%

Recent Advances in Embedding Methods for Multi-Object Tracking: A Survey

Wang¹,

Song²,

Hwang³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…In our last paper [1], we explained why we used our own synthetic datasets not other popular datasets like [19] [20]. Surprisingly soon later, we started to see booming growth in synthetic datasets such as MOTSynth-MOT-CVPR22 [21].…”

Section: Datasetsmentioning

confidence: 99%

People Tracking and Re-Identifying in Distributed Contexts: Extension Study of PoseTReID

Siv¹,

Mancaş²,

Gosselin³

et al. 2022

Preprint

View full text Add to dashboard Cite

In our previous paper, we introduced PoseTReID which is a generic framework for real-time 2D multi-person tracking in distributed interaction spaces where long-term people's identities are important for other studies such as behavior analysis, etc. In this paper, we introduce a further study of PoseTReID framework in order to give a more complete comprehension of the framework. We use a well-known bounding box detector YOLO (v4) for the detection to compare to OpenPose which was used in our last paper, and we use SORT and Deep-SORT to compare to centroid which was also used previously, and most importantly for the re-identification, we use a bunch of deep leaning methods such as MLFN, OSNet, and OSNet-AIN with our custom classification layer to compare to FaceNet which was also used earlier in our last paper. By evaluating on our PoseTReID datasets, even though those deep learning re-identification methods are designed for only short-term reidentification across multiple cameras or videos, it is worth showing that they give impressive results which boost the overall tracking performance of PoseTReID framework regardless the type of tracking method. At the same time, we also introduce our research-friendly and open source Python toolbox pyppbox, which is purely written in Python and contains all sub-modules which are used in this study along with real-time online and offline evaluations for our PoseTReID datasets. This pyppbox is available on GitHub https://github.com/rathaumons/pyppbox .

show abstract

“…Furthermore, recording personal visual data has brought privacy concerns (Akinosho et al, 2020). Ethical issues regarding privacy are of concern in many countries, which resulted in the passing of strict laws to protect privacy (Fabbri et al, 2021). In light of the recent General Data Protection Regulation (GDPR), which applies to all companies holding EU citizen data, care must be taken with data sources (Koops, 2014).…”

Section: Introductionmentioning

confidence: 99%

“…Ethical issues regarding privacy are also critical in the US. As an instance of privacy concerns, datasets for re-identification modules of DukeMTMC were taken offline recently (Fabbri et al, 2021;Ristani et al, 2016).…”

Section: Introductionmentioning

confidence: 99%

Fake it till you make it: Training Deep Neural Networks for Worker Detection using Synthetic Data

Tohidifar¹,

Saffari²,

Kim³

2022

Proceedings of the 29th EG-ICE International Workshop on Intelligent Computing in Engineering

View full text Add to dashboard Cite

The construction industry’s productivity and safety have long been a source of concern, while the broad use of deep neural network (DNN)-based visual AI has transformed other industries. Automation and digitalization powered by DNN provide intriguing answers; yetthe lack of high-quality, diversifieddataprevents the construction sector from leveragingthe benefits. This paper presentsa novel computational framework that enables synthetic data generationfor DNN training to overcome the time-consuming manual data collectionand avoiddata privacy problems. The suggested framework uses graphics engines to create a virtual duplicate of the constructionsite that generates non-real yet realistic visuals. The proposedframework randomizes crucial scene elements such as worker pose, clothes, camera viewpoint, and lighting conditionsto enhancethe variety of the synthetic dataset.The findings of this study presentpromisingpotential of synthetic datain DNN training.

show abstract

MOTSynth: How Can Synthetic Data Help Pedestrian Detection and Tracking?

Cited by 87 publications

References 54 publications

Recent Advances in Embedding Methods for Multi-Object Tracking: A Survey

Recent Advances in Embedding Methods for Multi-Object Tracking: A Survey

People Tracking and Re-Identifying in Distributed Contexts: Extension Study of PoseTReID

Fake it till you make it: Training Deep Neural Networks for Worker Detection using Synthetic Data

Contact Info

Product

Resources

About